4. DPDK Release 22.11

4.1. New Features

  • Added initial LoongArch architecture support.

    Added EAL implementation for LoongArch architecture. The initial port was tested on Loongson 3A5000, Loongson 3C5000 and Loongson 3C5000L. In theory this implementation should work with any target based on LoongArch ISA.

  • Added support for multiple mbuf pools per ethdev Rx queue.

    Added a capability which allows an application to provide many mempools of different size, and PMDs and/or NICs to choose a memory pool based on the packet’s length and/or Rx buffer availability.

  • Added support for congestion management in ethdev.

    Added new API functions rte_eth_cman_config_init(), rte_eth_cman_config_get(), rte_eth_cman_config_set() and rte_eth_cman_info_get() to support congestion management.

  • Added protocol header based buffer split.

    • Added rte_eth_buffer_split_get_supported_hdr_ptypes() to get supported header protocols to split at.

    • Added support for protocol-based buffer split using new proto_hdr field in structure rte_eth_rxseg_split.

  • Added proactive error handling mode for ethdev.

    Added proactive error handling mode for ethdev, and introduced three new events: RTE_ETH_EVENT_ERR_RECOVERING to report that the port is recovering from an error, RTE_ETH_EVENT_RECOVER_SUCCESS and RTE_ETH_EVENT_RECOVER_FAILED.

  • Added ethdev Rx/Tx descriptor dump API.

    Added the ethdev Rx/Tx descriptor dump API which provides functions for querying descriptor from device. The descriptor information differs in different NICs. The information demonstrates I/O process which is important for debug. The dump format is vendor-specific.

  • Added ethdev hairpin memory configuration options.

    Added new configuration flags for hairpin queues in rte_eth_hairpin_conf:

    • use_locked_device_memory

    • use_rte_memory

    • force_memory

    Each flag has a corresponding capability flag in struct rte_eth_hairpin_queue_cap.

  • Added strict queue to pre-configuration flow hints.

    • Added flags option to rte_flow_configure and rte_flow_info_get.

    • Added RTE_FLOW_PORT_FLAG_STRICT_QUEUE flag to indicate all operations for a given flow rule will strictly happen on the same flow queue.

  • Added configuration for asynchronous flow connection tracking.

    Added connection tracking action number hint to rte_flow_configure and rte_flow_info_get. The PMD can prepare the connection tracking resources according to the hint.

  • Added support for queue-based async query in flow API.

    Added new function rte_flow_async_action_handle_query() to query the action asynchronously.

  • Extended metering and marking support in the flow API.

    • Added METER_COLOR item to match color marker set by a meter.

    • Added ability to set color marker via modify field flow API.

    • Added meter API to get a pointer to the profile/policy by their ID.

    • Added METER_MARK action for metering with lockless profile/policy access.

  • Added flow offload action to route packets to kernel.

    Added new flow action which allows an application to re-route packets directly to the kernel without software involvement.

  • Updated AF_XDP driver.

    • Updated AF_XDP driver to make it compatible with libbpf v0.8.0 (when used with libxdp).

  • Updated AMD Pensando ionic driver.

    • Updated ionic driver to reflect that Pensando has been acquired by AMD.

    • Enhanced data path to provide substantial performance improvements.

    • Added support for mbuf fast free.

    • Added support for advertising packet types.

    • Added support for descriptor status functions.

    • Added Q-in-CMB feature controlled by device option ionic_cmb.

    • Added optimized handlers for non-scattered Rx and Tx.

  • Added GVE net PMD.

    • Added the new gve net driver for Google Virtual Ethernet devices.

    • See the GVE poll mode driver NIC guide for more details on this new driver.

  • Updated Intel iavf driver.

    • Added flow subscription support.

  • Updated Intel ice driver.

    • Added protocol based buffer split support in scalar path.

  • Added Intel idpf driver.

    Added the new idpf net driver for Intel® Infrastructure Processing Unit (Intel® IPU) E2100. See the IDPF Poll Mode Driver NIC guide for more details on this new driver.

  • Updated Marvell cnxk driver.

    • Added support for flow action REPRESENTED_PORT.

    • Added support for congestion management.

  • Added Microsoft mana driver.

    The driver has been disabled by default because of a, currently, missing dependency.

  • Updated Netronome nfp driver.

    Added flow API support:

    • Added support for the flower firmware.

    • Added the flower service infrastructure.

    • Added the control message interactive channels between PMD and firmware.

    • Added support for a representor port.

  • Updated NVIDIA mlx5 driver.

    • Added full support for queue-based async hardware steering.

      • Support of FDB.

      • Support of control flow and isolate mode.

      • Support of conntrack.

      • Support of counter.

      • Support of aging.

      • Support of meter.

      • Support of modify fields.

  • Updated NXP dpaa2 driver.

    • Added support for flow action REPRESENTED_PORT.

  • Updated Wangxun ngbe driver.

    • Added support to set device link down/up.

  • Added DMA vChannel unconfiguration for async vhost.

    Added support to unconfigure DMA vChannel that is no longer used by the vhost library.

  • Added non-blocking notify API to vhost library.

    Added rte_vhost_vring_call_nonblock API to notify the guest that used descriptors have been added to the vring in n aon-blocking way. The user should check the return value of this API and try again if needed.

  • Added support for MACsec in rte_security.

    Added MACsec transform for rte_security session and added new API to configure security associations (SA) and secure channels (SC).

  • Added new algorithms to cryptodev.

    • Added symmetric hash algorithm ShangMi 3 (SM3).

    • Added symmetric cipher algorithm ShangMi 4 (SM4) in ECB, CBC and CTR modes.

  • Updated Intel QuickAssist Technology (QAT) crypto driver.

    • Added support for SM3 hash algorithm.

    • Added support for SM4 encryption algorithm in ECB, CBC and CTR modes.

    • Added support for ECDH key exchange algorithm.

  • Updated Marvell cnxk crypto driver.

    • Added AES-CCM support in lookaside protocol (IPsec) for CN9K and CN10K.

    • Added AES & DES DOCSIS algorithm support in lookaside crypto for CN9K.

  • Updated aesni_mb crypto driver.

    • Added support for 8-byte and 16-byte tags for ZUC-EIA3-256.

    • Added support for in-place SGL, out-of-place SGL in SGL out, out-of-place LB in SGL out, and out-of-place SGL in LB out.

  • Updated ipsec_mb crypto driver.

    • Added SNOW-3G and ZUC support for ARM platform.

    • Added Intel IPsec MB v1.3 library support for x86 platform.

  • Added UADK crypto driver.

    Added a new crypto driver for the UADK library. See the UADK Crypto Poll Mode Driver guide for more details on this new driver.

  • Added bbdev operation for FFT processing.

    Added a new operation type in bbdev for FFT processing with new functions rte_bbdev_enqueue_fft_ops and rte_bbdev_dequeue_fft_ops, and related structures.

  • Added Intel ACC200 bbdev driver.

    Added a new acc200 bbdev driver for the Intel® ACC200 accelerator integrated on SPR-EE. See the bbdevs/acc200 guide for more details on this new driver.

  • Added eventdev adapter instance get API.

    • Added rte_event_eth_rx_adapter_instance_get to get Rx adapter instance ID for specified ethernet device ID and Rx queue index.

    • Added rte_event_eth_tx_adapter_instance_get to get Tx adapter instance ID for specified ethernet device ID and Tx queue index.

  • Added eventdev Tx adapter queue start/stop API.

    • Added rte_event_eth_tx_adapter_queue_start to start enqueueing packets to the Tx queue by Tx adapter.

    • Added rte_event_eth_tx_adapter_queue_stop to stop the Tx Adapter from enqueueing any packets to the Tx queue.

  • Added event crypto adapter vectorization support.

    Added support for aggregating crypto operations processed by event crypto adapter into a single event containing rte_event_vector whose event type is RTE_EVENT_TYPE_CRYPTODEV_VECTOR.

  • Added NitroSketch in membership library.

    Added a new data structure called sketch into the membership library, to profile the traffic efficiently. NitroSketch provides high-fidelity approximate measurements and appears as a promising alternative to traditional approaches such as packet sampling.

  • Added Intel uncore frequency control API to the power library.

    Added API to allow uncore frequency adjustment. This is done through manipulating related uncore frequency control sysfs entries to adjust the minimum and maximum uncore frequency values, which works on Linux with Intel hardware only.

  • Added security performance test application.

    Added new application to test rte_security session create/destroy performance. See the Security Performance Tool for more details.

  • Updated IPsec sample application.

    Added support for lookaside sessions in event mode. See the IPsec Security Gateway Sample Application for more details.

  • Updated FIPS validation sample application.

    Added support for asymmetric crypto algorithms. See the Federal Information Processing Standards (FIPS) CryptoDev Validation for more details.

  • Rewrote pmdinfo script.

    The dpdk-pmdinfo.py script was rewritten to produce valid JSON only. PCI-IDs parsing has been removed. To get a similar output to the (now removed) -r/--raw flag, the following command may be used:

    strings $dpdk_binary_or_driver | sed -n 's/^PMD_INFO_STRING= //p'

4.2. Removed Items

  • mem: Removed not implemented and deprecated rte_malloc_set_limit.

  • ethdev: removed RTE_FLOW_ITEM_TYPE_PF; use RTE_FLOW_ITEM_TYPE_REPRESENTED_PORT instead.

  • ethdev: removed RTE_FLOW_ITEM_TYPE_VF; use RTE_FLOW_ITEM_TYPE_REPRESENTED_PORT instead.

  • ethdev: removed RTE_FLOW_ITEM_TYPE_PHY_PORT; use RTE_FLOW_ITEM_TYPE_REPRESENTED_PORT instead.

  • ethdev: removed RTE_FLOW_ACTION_TYPE_PHY_PORT; use RTE_FLOW_ACTION_TYPE_REPRESENTED_PORT instead.

  • ethdev: removed OF_SET_MPLS_TTL, OF_DEC_MPLS_TTL, OF_SET_NW_TTL, OF_COPY_TTL_OUT and OF_COPY_TTL_IN which are not actually supported by any PMD. MODIFY_FIELD action should be used to do packet edits via flow API.

  • vhost: Removed deprecated rte_vhost_gpa_to_vva and rte_vhost_get_queue_num helpers.

4.3. API Changes

  • eal: RTE_FUNC_PTR_OR_* macros have been marked deprecated and will be removed in the future. Applications can use devtools/cocci/func_or_ret.cocci to update their code.

  • eal: Updated rte_eal_remote_launch so it returns -EPIPE in case of a read or write error on the pipe, instead of calling rte_panic.

  • eal: Updated return types for rte_{bsf,fls} inline functions to be consistently uint32_t.

  • mempool: Deprecated helper macro MEMPOOL_HEADER_SIZE() has been removed. The replacement macro RTE_MEMPOOL_HEADER_SIZE() is internal only.

  • mempool: Deprecated macro to register mempool driver MEMPOOL_REGISTER_OPS() has been removed. Use replacement macro RTE_MEMPOOL_REGISTER_OPS() instead.

  • mempool: Deprecated macros MEMPOOL_PG_NUM_DEFAULT and MEMPOOL_PG_SHIFT_MAX have been removed. These macros are not used and not required any more.

  • mbuf: Removed deprecated PKT_* flags. Use corresponding flags with RTE_MBUF_F_ prefix instead. Applications can use devtools/cocci/prefix_mbuf_offload_flags.cocci to replace all occurrences of old mbuf flags in C code.

  • bus: Changed the device numa node to -1 when NUMA information is unavailable. The dev->device.numa_node field is set by each bus driver for every device it manages to indicate on which NUMA node this device lies. When this information is unknown, the assigned value was not consistent across the bus drivers. This similarly impacts rte_eth_dev_socket_id().

  • bus: Registering a bus has been marked as an internal API. External users may still register their bus using the bus_driver.h driver header (see enable_driver_sdk meson option). The rte_bus object is now opaque and must be manipulated through added accessors.

  • drivers: Registering a driver on the auxiliary, ifpga, pci, vdev, vmbus buses has been marked as an internal API. External users may still register their driver using the associated driver headers (see enable_driver_sdk meson option). The rte_driver and rte_device objects are now opaque and must be manipulated through added accessors.

  • ethdev: Removed deprecated macros. Applications can use devtools/cocci/namespace_ethdev.cocci to update their code.

    • Removed deprecated ETH_LINK_SPEED_*, ETH_SPEED_NUM_* and ETH_LINK_* (duplex-related) defines. Use corresponding defines with RTE_ prefix instead.

    • Removed deprecated ETH_MQ_RX_* and ETH_MQ_TX_* defines. Use corresponding defines with RTE_ prefix instead.

    • Removed deprecated ETH_RSS_* defines for hash function and RETA size specification. Use corresponding defines with RTE_ prefix instead.

    • Removed deprecated DEV_RX_OFFLOAD_* and DEV_TX_OFFLOAD_ defines. Use corresponding defines with RTE_ETH_RX_OFFLOAD_ and RTE_ETH_TX_OFFLOAD_ prefix instead.

    • Removed deprecated ETH_DCB_*, ETH_VMDQ_, ETH_*_TCS, ETH_*_POOLS and ETH_MAX_VMDQ_POOL defines. Use corresponding defines with RTE_ prefix instead.

    • Removed deprecated RTE_TUNNEL_* defines. Use corresponding defines with RTE_ETH_TUNNEL_ prefix instead.

    • Removed deprecated RTE_FC_* defines. Use corresponding defines with RTE_ETH_FC_ prefix instead.

    • Removed deprecated ETH_VLAN_* and ETH_QINQ_ defines. Use corresponding defines with RTE_ prefix instead.

    • Removed deprecated ETH_NUM_RECEIVE_MAC_ADDR define. Use corresponding define with RTE_ prefix instead.

    • Removed deprecated PKT_{R,T}X_DYNF_METADATA defines. Use corresponding defines RTE_MBUF_DYNFLAG_{R,T}X_METADATA instead.

  • ethdev: Removed deprecated Flow Director configuration from device configuration (dev_conf.fdir_conf). Moved corresponding structures to internal API since some drivers still use it internally.

  • ethdev: Removed the Rx offload flag RTE_ETH_RX_OFFLOAD_HEADER_SPLIT and field split_hdr_size from the structure rte_eth_rxmode used to configure header split. Instead, user can still use RTE_ETH_RX_OFFLOAD_BUFFER_SPLIT for per-queue packet split offload, which is configured by rte_eth_rxseg_split.

  • ethdev: The reserved field in the rte_eth_rxseg_split structure is replaced with proto_hdr to support protocol header based buffer split. User can choose length or protocol header to configure buffer split according to NIC’s capability.

  • ethdev: Changed the type of the parameter rate of the function rte_eth_set_queue_rate_limit() from uint16_t to uint32_t to support more than 64 Gbps. Changed the type of the parameter tx_rate of the functions rte_pmd_bnxt_set_vf_rate_limit() and rte_pmd_ixgbe_set_vf_rate_limit() in the same way for consistency.

  • ethdev: Promoted rte_eth_rx_metadata_negotiate() from experimental to stable.

  • ethdev: Promoted the following flow primitives from experimental to stable:

    • RTE_FLOW_ACTION_TYPE_PORT_REPRESENTOR

    • RTE_FLOW_ACTION_TYPE_REPRESENTED_PORT

    • RTE_FLOW_ITEM_TYPE_PORT_REPRESENTOR

    • RTE_FLOW_ITEM_TYPE_REPRESENTED_PORT

  • ethdev: Promoted rte_flow_pick_transfer_proxy() from experimental to stable.

  • ethdev: Banned the use of attributes ingress/egress in “transfer” flows, as the final step of the deprecation process that had been started in DPDK 21.11. See items PORT_REPRESENTOR, REPRESENTED_PORT.

  • vhost: Promoted rte_vhost_vring_stats_get(), rte_vhost_vring_stats_get_names() and rte_vhost_vring_stats_reset() from experimental to stable.

  • cryptodev: The structure rte_cryptodev_sym_session was made internal. The API rte_cryptodev_sym_session_init and rte_cryptodev_sym_session_clear were removed and user would only need to call rte_cryptodev_sym_session_create and rte_cryptodev_sym_session_free to create/destroy sessions. The API rte_cryptodev_sym_session_create was updated to take a single mempool with element size big enough to hold session data and session private data. All sample applications were updated to attach an opaque pointer for the session to the rte_crypto_op while enqueuing.

  • security: The structure rte_security_session was made internal and corresponding functions were updated to take/return an opaque session pointer. The API rte_security_session_create was updated to take only one mempool which has enough space to hold session and driver private data.

  • security: MACsec support has been added which resulted in updates to structures rte_security_macsec_xform, rte_security_macsec_stats and security capability structure rte_security_capability to accommodate MACsec capabilities.

  • security: The experimental API rte_security_get_userdata was being unused by most of the drivers and it was retrieving userdata from mbuf dynamic field. The API is now removed and the application can directly get the userdata from mbuf dynamic field.

  • eventdev: The function rte_event_crypto_adapter_queue_pair_add was updated to accept configuration of type rte_event_crypto_adapter_queue_conf instead of rte_event, similar to rte_event_eth_rx_adapter_queue_add signature. Event will be one of the configuration fields, together with additional vector parameters.

  • eventdev: The function pointer definition eventdev_stop_flush_t is renamed to rte_eventdev_stop_flush_t to avoid conflicts with application symbols.

  • eventdev: The data type of the ID parameter in the functions rte_event_dev_xstats_names_get, rte_event_dev_xstats_get, rte_event_dev_xstats_by_name_get and rte_event_dev_xstats_reset is changed to uint64_t from unsigned int and uint32_t.

  • metrics: Updated rte_metrics_init so it returns an error code instead of calling rte_exit.

  • telemetry: The allowed characters in names for dictionary values are now limited to alphanumeric characters and a small subset of additional printable characters. This will ensure that all dictionary parameter names can be output without escaping in JSON - or in any future output format used. Names for the telemetry commands are now similarly limited. The parameters for telemetry commands are unaffected by this change.

  • raw/ifgpa: The function rte_pmd_ifpga_get_pci_bus has been removed.

4.4. ABI Changes

  • eal: Updated EAL thread names from lcore-worker-<lcore_id> to rte-worker-<lcore_id> so that DPDK can accommodate lcores higher than 99.

  • mbuf: Replaced buf_iova field with next field and added a new field dynfield2 at its place in second cacheline if RTE_IOVA_AS_PA is 0.

  • ethdev: enum RTE_FLOW_ITEM was affected by deprecation procedure.

  • ethdev: enum RTE_FLOW_ACTION was affected by deprecation procedure.

  • bbdev: enum rte_bbdev_op_type was affected to remove RTE_BBDEV_OP_TYPE_COUNT and to allow for futureproof enum insertion a padded RTE_BBDEV_OP_TYPE_SIZE_MAX macro is added.

  • bbdev: Structure rte_bbdev_driver_info was updated to add new parameters for queue topology, device status using rte_bbdev_device_status.

  • bbdev: Structure rte_bbdev_queue_data was updated to add new parameter for enqueue status using rte_bbdev_enqueue_status.

  • eventdev: Added evtim_drop_count field to rte_event_timer_adapter_stats structure.

  • eventdev: Added weight and affinity fields to rte_event_queue_conf structure.

  • eventdev: The field *u64s in the structure rte_event_vector is replaced with u64s as the field is supposed to hold an array of uint64_t values.

  • eventdev: The structure rte_event_vector was updated to include a new bit field elem_offset:12. The bits are taken from the bitfield rsvd:15.

4.5. Tested Platforms

  • Intel® platforms with Intel® NICs combinations

    • CPU

      • Intel® Atom™ CPU C3758 @ 2.20GHz

      • Intel® Xeon® CPU D-1553N @ 2.30GHz

      • Intel® Xeon® CPU E5-2680 v2 @ 2.80GHz

      • Intel® Xeon® CPU E5-2699 v3 @ 2.30GHz

      • Intel® Xeon® CPU E5-2699 v4 @ 2.20GHz

      • Intel® Xeon® D-2796NT CPU @ 2.00GHz

      • Intel® Xeon® Gold 6139 CPU @ 2.30GHz

      • Intel® Xeon® Gold 6140M CPU @ 2.30GHz

      • Intel® Xeon® Gold 6252N CPU @ 2.30GHz

      • Intel® Xeon® Gold 6348 CPU @ 2.60GHz

      • Intel® Xeon® Platinum 8180M CPU @ 2.50GHz

      • Intel® Xeon® Platinum 8280M CPU @ 2.70GHz

      • Intel® Xeon® Platinum 8380 CPU @ 2.30GHz

    • OS:

      • Fedora 36

      • FreeBSD 13.1

      • Red Hat Enterprise Linux Server release 8.6

      • Red Hat Enterprise Linux Server release 9

      • CentOS 7.9

      • Ubuntu 20.04.5

      • Ubuntu 22.04.1

      • Ubuntu 22.10

      • SUSE Linux Enterprise Server 15 SP4

    • NICs:

      • Intel® Ethernet Controller E810-C for SFP (4x25G)

        • Firmware version: 4.10 0x800151d8 1.3310.0

        • Device id (pf/vf): 8086:1593 / 8086:1889

        • Driver version(out-tree): 1.10.6 (ice)

        • Driver version(in-tree): 5.15.0-46-generic / 4.18.0-372.9.1.rt7.166.el8.x86_64 (ice)

        • OS Default DDP: 1.3.30.0

        • COMMS DDP: 1.3.37.0

        • Wireless Edge DDP: 1.3.10.0

      • Intel® Ethernet Controller E810-C for QSFP (2x100G)

        • Firmware version: 4.10 0x8001518e 1.3310.0

        • Device id (pf/vf): 8086:1592 / 8086:1889

        • Driver version: 1.10.6 (ice)

        • OS Default DDP: 1.3.30.0

        • COMMS DDP: 1.3.37.0

        • Wireless Edge DDP: 1.3.10.0

      • Intel® Ethernet Controller E810-XXV for SFP (2x25G)

        • Firmware version: 4.10 0x80015188 1.3310.0

        • Device id (pf/vf): 8086:159b / 8086:1889

        • Driver version: 1.10.6 (ice)

        • OS Default DDP: 1.3.30.0

        • COMMS DDP: 1.3.37.0

      • Intel® 82599ES 10 Gigabit Ethernet Controller

        • Firmware version: 0x61bf0001

        • Device id (pf/vf): 8086:10fb / 8086:10ed

        • Driver version(out-tree): 5.16.5 (ixgbe)

        • Driver version(in-tree): 5.15.0-46-generic (ixgbe)

      • Intel® Ethernet Converged Network Adapter X710-DA4 (4x10G)

        • Firmware version: 9.00 0x8000cead 1.3179.0

        • Device id (pf/vf): 8086:1572 / 8086:154c

        • Driver version(out-tree): 2.20.12 (i40e)

        • Driver version(in-tree): 5.15.0-46-generic (i40e)

      • Intel® Corporation Ethernet Connection X722 for 10GbE SFP+ (2x10G)

        • Firmware version: 6.00 0x800039ec 1.3179.0

        • Device id (pf/vf): 8086:37d0 / 8086:37cd

        • Driver version(out-tree): 2.20.12 (i40e)

        • Driver version(in-tree): 5.15.0-46-generic (i40e)

      • Intel® Corporation Ethernet Connection X722 for 10GBASE-T

        • Firmware version: 6.00 0x800039aa 1.2935.0

        • Device id (pf/vf): 8086:37d2 / 8086:37cd

        • Driver version(out-tree): 2.20.12 (i40e)

        • Driver version(in-tree): 5.15.0-46-generic (i40e)

      • Intel® Ethernet Converged Network Adapter XXV710-DA2 (2x25G)

        • Firmware version: 9.00 0x8000ce90 1.3179.0

        • Device id (pf/vf): 8086:158b / 8086:154c

        • Driver version(out-tree): 2.20.12 (i40e)

        • Driver version(in-tree): 5.15.0-46-generic (i40e)

      • Intel® Ethernet Converged Network Adapter XL710-QDA2 (2X40G)

        • Firmware version(PF): 9.00 0x8000ce86 1.3179.0

        • Device id (pf/vf): 8086:1583 / 8086:154c

        • Driver version(out-tree): 2.20.12 (i40e)

        • Driver version(in-tree): 5.15.0-46-generic (i40e)

      • Intel® Ethernet Converged Network Adapter X710-T2L

        • Firmware version: 9.00 0x8000ce67 1.3179.0

        • Device id (pf): 8086:15ff

        • Driver version: 2.20.12 (i40e)

  • Intel® platforms with NVIDIA® NICs combinations

    • CPU:

      • Intel® Xeon® Gold 6154 CPU @ 3.00GHz

      • Intel® Xeon® CPU E5-2697A v4 @ 2.60GHz

      • Intel® Xeon® CPU E5-2697 v3 @ 2.60GHz

      • Intel® Xeon® CPU E5-2680 v2 @ 2.80GHz

      • Intel® Xeon® CPU E5-2670 0 @ 2.60GHz

      • Intel® Xeon® CPU E5-2650 v4 @ 2.20GHz

      • Intel® Xeon® CPU E5-2650 v3 @ 2.30GHz

      • Intel® Xeon® CPU E5-2640 @ 2.50GHz

      • Intel® Xeon® CPU E5-2650 0 @ 2.00GHz

      • Intel® Xeon® CPU E5-2620 v4 @ 2.10GHz

    • OS:

      • Red Hat Enterprise Linux release 8.6 (Ootpa)

      • Red Hat Enterprise Linux release 8.4 (Ootpa)

      • Red Hat Enterprise Linux release 8.2 (Ootpa)

      • Red Hat Enterprise Linux Server release 7.9 (Maipo)

      • Red Hat Enterprise Linux Server release 7.8 (Maipo)

      • Red Hat Enterprise Linux Server release 7.6 (Maipo)

      • Red Hat Enterprise Linux Server release 7.5 (Maipo)

      • Red Hat Enterprise Linux Server release 7.4 (Maipo)

      • Ubuntu 22.04

      • Ubuntu 20.04

      • Ubuntu 18.04

      • SUSE Enterprise Linux 15 SP2

    • OFED:

      • MLNX_OFED 5.8-1.0.1.1 and above

      • MLNX_OFED 5.7-1.0.2.0

    • upstream kernel:

      • Linux 6.1.0-rc3 and above

    • rdma-core:

      • rdma-core-43.0 and above

    • NICs:

      • NVIDIA® ConnectX®-3 Pro 40G MCX354A-FCC_Ax (2x40G)

        • Host interface: PCI Express 3.0 x8

        • Device ID: 15b3:1007

        • Firmware version: 2.42.5000

      • NVIDIA® ConnectX®-3 Pro 40G MCX354A-FCCT (2x40G)

        • Host interface: PCI Express 3.0 x8

        • Device ID: 15b3:1007

        • Firmware version: 2.42.5000

      • NVIDIA® ConnectX®-4 Lx 25G MCX4121A-ACAT (2x25G)

        • Host interface: PCI Express 3.0 x8

        • Device ID: 15b3:1015

        • Firmware version: 14.32.1010 and above

      • NVIDIA® ConnectX®-4 Lx 50G MCX4131A-GCAT (1x50G)

        • Host interface: PCI Express 3.0 x8

        • Device ID: 15b3:1015

        • Firmware version: 14.32.1010 and above

      • NVIDIA® ConnectX®-5 100G MCX516A-CCAT (2x100G)

        • Host interface: PCI Express 3.0 x16

        • Device ID: 15b3:1017

        • Firmware version: 16.35.1012 and above

      • NVIDIA® ConnectX®-5 100G MCX556A-ECAT (2x100G)

        • Host interface: PCI Express 3.0 x16

        • Device ID: 15b3:1017

        • Firmware version: 16.35.1012 and above

      • NVIDIA® ConnectX®-5 100G MCX556A-EDAT (2x100G)

        • Host interface: PCI Express 3.0 x16

        • Device ID: 15b3:1017

        • Firmware version: 16.35.1012 and above

      • NVIDIA® ConnectX®-5 Ex EN 100G MCX516A-CDAT (2x100G)

        • Host interface: PCI Express 4.0 x16

        • Device ID: 15b3:1019

        • Firmware version: 16.35.1012 and above

      • NVIDIA® ConnectX®-6 Dx EN 100G MCX623106AN-CDAT (2x100G)

        • Host interface: PCI Express 4.0 x16

        • Device ID: 15b3:101d

        • Firmware version: 22.35.1012 and above

      • NVIDIA® ConnectX®-6 Lx EN 25G MCX631102AN-ADAT (2x25G)

        • Host interface: PCI Express 4.0 x8

        • Device ID: 15b3:101f

        • Firmware version: 26.35.1012 and above

      • NVIDIA® ConnectX®-7 200G CX713106AE-HEA_QP1_Ax (2x200G)

        • Host interface: PCI Express 5.0 x16

        • Device ID: 15b3:1021

        • Firmware version: 28.35.1012 and above

  • NVIDIA® BlueField® SmartNIC

    • NVIDIA® BlueField®-2 SmartNIC MT41686 - MBF2H332A-AEEOT_A1 (2x25G)

      • Host interface: PCI Express 3.0 x16

      • Device ID: 15b3:a2d6

      • Firmware version: 24.35.1012 and above

    • Embedded software:

      • Ubuntu 20.04.3

      • MLNX_OFED 5.8-1.0.1.1 and above

      • DOCA 1.5 with BlueField 3.9.3

      • DPDK application running on ARM cores

  • IBM Power 9 platforms with NVIDIA® NICs combinations

    • CPU:

      • POWER9 2.2 (pvr 004e 1202)

    • OS:

      • Ubuntu 20.04

    • NICs:

      • NVIDIA® ConnectX®-5 100G MCX556A-ECAT (2x100G)

        • Host interface: PCI Express 4.0 x16

        • Device ID: 15b3:1017

        • Firmware version: 16.35.1012

      • NVIDIA® ConnectX®-6 Dx 100G MCX623106AN-CDAT (2x100G)

        • Host interface: PCI Express 4.0 x16

        • Device ID: 15b3:101d

        • Firmware version: 22.35.1012

    • OFED:

      • MLNX_OFED 5.8-1.0.1.1